
% Author: Kazuma Takakura
% Date: 17 May 2025
% Time: 10:53:19

\begin{table}[h!]\footnotesize
  \centering
  \caption{Inverse Probability Weighted Difference in Differences}
\label{tab:did_ipw}
\scalebox{1}{
\begin{threeparttable}
\begin{tabular}{lccc}\toprule
  & Rapid math test score &  RSES score  & CPCS score   \\\midrule\midrule
  DiD & -0.323  & 0.586* & 0.596* \\ 
    & (0.232) & (0.343) & (0.333) \\ 
  Treatment & 0.044  & -0.056 & -0.034 \\ 
    & (0.180) & (0.157) & (0.147) \\ 
  After & 0.200  & -0.356 & -0.357 \\ 
    & (0.193) & (0.244) & (0.229) \\ 
  Constant & 0.209*  & -0.001 & -0.025 \\ 
    & (0.122) & (0.102) & (0.095) \\ 
  Observations & 243  & 236 & 236 \\ 
\midrule
\end{tabular}
\begin{tablenotes}
\item (a) We control student's grade, sex, baseline cognitive and baseline non-cognitive score, DT baseline time, branch dummy (location), parents' income source, last income per family member, number of household members, age of household head, education level of household head, teacher's age, sex, and phone survey dummy. The same variables are used for propensity score calculation.
\item (b) School-clustered standard errors are reported within parentheses. 
\item (c) $^*$ Significant at 10\% level; $^{**}$ significant at 5\% level; $^{***}$ significant at 1\% level. 
\end{tablenotes}
\end{threeparttable}
}
\label{tab:addlabel}%
\end{table}




